Statistical Analysis of Randomized Experiments with Nonignorable Missing Outcomes

• Activity Number: 234
• Type: Topic Contributed
• Date/Time: Tuesday, July 31, 2007 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): Kosuke Imai*+
• Companies: Princeton University
• Keywords: causal inference ; noncompliance ; instrumental variables ; average treatment effects ; sensitivity analysis ; identification

Abstract

Missing data are frequently encountered in the statistical analysis of randomized experiments. In this paper, I propose statistical methods that can be used to analyze randomized experiments with a nonignorable missing binary outcome where the missing-data mechanism may depend on the unobserved values of the outcome variable itself. I first introduce new identification strategies for the average treatment effects and complier average causal effects. I then derive the maximum likelihood estimator and its asymptotic properties, and discuss possible estimation methods. Since the proposed identification assumption is not directly verifiable from the data, I show how to conduct a sensitivity analysis based on the parameterization that links the key identification assumption with the causal quantities of interest. I apply the proposed methods to analyze data from two randomized experiments.